The need for a safe and effective bone substitute still exists in dentistry and orthopedics: bone graft materials are difficult to work with and allografts may be unsafe; the rate of new bone growth associated with bone grafts is unacceptably slow; the long-term stability of bone graft substitutes is doubtful and osteoconduction is unacceptably slow; and the promise of a graft material (or graft substitute) augmented with a biological substance (e.g., osteogenic cytokine) has yet to be realized. A delivery system (GAM) has been developed in which osteoinductive plasmid genes are delivered from a moldable, biodegradable structural matrix. Successful feasibility studies have been conducted in rat and canine models, and this work has shown for the first time that new bone will form rapidly in vivo following direct osteoinductive plasmid gene transfer to fibroblasts involved in skeletal repair. The feasibility studies support an overall hypothesis that rapid new bone formation can be induced via GAM osteoinductive plasmid gene transfer. However, a consistent finding has been that the center of a large osseous defect is the last and most difficult region to regenerate. Because full-thickness regeneration is crucial to clinical success, this grant application proposes a series of interdisciplinary experiments that are designed to understand the mechanism of GAM gene transfer. In essence, we believe that full- thickness regeneration of large osteotomy defects will only be realized through an increased understanding of the basic mechanisms associated with plasmid gene transfer. The specific hypothesis to be explored in this application is that GAM constructs can be pre-designed so as to increase the percentage of genetically modified fibroblasts and, therefore, the rate and amount of new bone that forms in vivo. In this regard, a series of rational modifications to the GAM plasmid DNA and the GAM structural matrix are proposed (Specific Aims 1-2). By studying these modifications, important new insights into the mechanism of GAM gene transfer will be gained. How these modifications affect the behavior of osteogenic cells following gene transfer will be measured in a cell culture model system in vitro (Specific Aim 3). From these measures we should gain new insight into the biological process of new bone formation. While likely beyond the scope of the present application, a long term goal will be to measure the efficacy of rationally pre-designed GAM constructs in vivo using established bone defect animal models.